The present invention relates to a method and a device for the direct reduction of iron ore with coal.
Such methods and devices are employed in the production of iron or steel from iron ore, ie in metallurgy for example.
The methods applied for by far the greatest part of iron production from ore utilise the carbon in coal or coke for two processes which proceed simultaneously.
Carbon acts first of all as a chemical reducing agent and takes over the oxygen from the iron oxide.
Secondly, by combustion with oxygen (usually from the air) carbon supplies the necessary heat for the process.
For the best utilisation of the coal each of these two process steps requires thermodynamic conditions which are mutually exclusive (xe2x80x9cthermodynamic dilemmaxe2x80x9d).
For the reduction, the chemical potential of the oxygen in the atmosphere must be as low as possible, ie high CO content, low CO2 content and low O2 content in equilibrium with CO.
For the most complete possible combustion of the carbon to CO2, an O2 content in the combustion atmosphere meeting at least the stoichiometric requirements is needed.
In the coal reduction methods currently used in industry compromises are entered into. Accordingly, more energy is expended than is needed for the actual iron ore production process. At the same time the methods result either in reduction gas which does not have the full capability for reducing the oxide because before it is used for reduction the gas has already taken up oxygen by partial combustion or they result in the heat needed not being produced with optimum efficiency since the gas cannot be burned completely to CO2.
Depending on the process, the unused portion of energy is dissipated in the form of carbon monoxide and hydrogen or as perceptible heat or both. The more or less advantageous use of the coupled product, energy, from these non-autothermic processes has a substantial effect on the economic efficiency of the overall process. In the blast furnace method the excess energy is given off in chemical form as furnace gas usually to usage loads located in the works. In the case of the rotary-tube furnace method the utilisation of the excess energy is difficult because frequently no usage loads are available. Occasionally the excess energy is converted to electrical energy.
In order to improve the energy yield in the context addressed here two routes have been taken. The compromise referred to has been improved or processes have been proposed which proceed without excess energy and are thus autothermic.
The characteristics of known methods affecting energy demand are here examined in more detail and compared.
For this purpose the RIST diagram in FIG. 1 is used. This shows on the abscissa the oxidation of the carbon as the molar ratio from O/C=0 to O/C=2, ie from carbon C to carbon dioxide CO2, and on the ordinate the origin of the required oxygen with reference to one mol of Fe and thus expressed by the molar ratio O/Fe.
This oxygen taken up by the carbon can come from the ore (central region of the ordinate) or it can be free oxygen for precombustion (lower part) or afterburning (upper part). The terms precombustion and afterburning refer to the combustion of C or CO before and after, respectively, these have been used for reduction.
The removal of oxygen from the oxide illustrating the reduction becomes more clearly discernible if the number of mols of oxygen present in the oxide in question during the reduction process is always related tob 1 mol of iron instead of the conventional 1, 2 or 3 mols of iron. The following notations then emerge:
haematite, Fe2O3, becomes FeO1.5;
magnetite, Fe3O4, becomes FeO1.33; and
wxc3xcstite, FeO, becomes FeO1.05.
The quantity of oxygen required for the combustion of a portion of the coal or CO (precombustion/afterburning) is also related to 1 mol of iron. Precombustion is given a negative sign and is plotted to xe2x88x921.5. If, for example, 1.5 mols of oxygen/mol of Fe are needed for precombustion the diagram shows that just as much oxygen for precombustion as for reduction, ie 1.5 mol O per mol Fe in each case and thus 3 O/Fe in total, is removed from the reactor with the flue gas.
A to D are process lines for different reduction processes. The gradient of a process line, that is the quotient O/Fe/O/C, abbreviates to C/Fe the specific carbon consumption. The latter is particularly important for the deliberations pursued here and is specified in the legend of the figure for the methods under consideration both in terms of mol C/mol Fe and kg C/t Fe.
Process line B represents the exchange of oxygen between iron oxide and carbon as well as the further take up of oxygen by the carbon in afterburning for an ideal autothermic process for the production of directly reduced iron (DRI). The carbon demand is 1.4 mol C per mol Fe or 299 kg C/t Fe and is thus almost the lowest possible for a method based only on carbon for reduction and heat generation. The diagram shows that the reduction of the iron oxide oxidises the carbon completely to CO and beyond that converts 10% of it to CO2. The remaining gas containing 90% of CO is afterburnt and the heat liberated covers the remaining enthalpy requirement of the reaction.
Process line A represents the rotary-tube method. The energy input required is greater than that for an autothermic process as can be seen from comparison of the gradients of the process lines. The actual energy demand for reduction in the rotary tube can naturally not be greater than that of an autothermic process but as a consequence of the process substantially more carbon must be fed in, that is to say 2.1 mol C/mol Fe, ie 451 kg C/t Fe, so that 152 kg C/t Fe are converted into heat lost in the process. The reason for this is that the CO formed in the reduction reaction must be burnt over the bed of coal and ore in order to generate the heat required for the Boudouard reaction. This combustion reaction, however, cannot be carried out completely to CO2 because the CO2 would impair the reduction process. It follows from this that any supply of heat without material separation of the gases in the reduction chamber and combustion chamber results for theoretical reasons in a consumption of carbon and energy which must be higher than that of a hypothetical autothermic process.
Process line C represents the oxygen exchange in a blast furnace operating only with coke. Due to the production of liquid end products as opposed to solid end products in processes A and B the consumption values have only limited comparability. In the blast furnace both precombustion (in the tuyere level) and afterburning (in the air heater) are employed. In the calculation of the carbon consumption the 0.2 C/Fe which leaves the furnace in the pig iron containing 4.3% C have already be subtracted.
Due to the partial combustion of the carbon in the tuyere level (precombustion) to produce the enthalpy needed for melting and heating the charge and for the Boudouard reaction the starting point of the process line is lowered by comparison with processes A and B from the origin of the coordinate system to xe2x88x921.30 O/Fe. Due to this preliminary oxidation of the carbon the gas in the upper section of the blast furnace contains more CO2 than originates from the reduction reaction. The gas is thus xe2x80x9cdilutedxe2x80x9d with oxygen and is no longer as well suited for reducing as it would be without this take-up of oxygen. Accordingly, in contrast with A and B the process line C passes close to the xe2x80x9cforbidden regionxe2x80x9d of the diagram marked by W and M. Since the compositions of the reducing gas lying in this region no longer result in the formation of metallic iron a process line cannot run through this zone.
Since the carbon consumption of the blast furnace is higher than that of an autothermic process possibilities for lowering the consumption could be sought. This would mean that the slope of the process line is reduced. However, it can immediately be seen from the diagram that this would only be possible if the point of intersection of the process line with the ordinate could be shifted upwards as a flatter process line would otherwise run into the xe2x80x9cforbidden zonexe2x80x9d. This point of intersection, however, is fixed by the oxygen demand for precombustion and this in turn results from the heat demand for fusion and the other processes requiring heat in the lower part of the furnace. Since, however, the heat demand in the lower part of the furnace can only be reduced very slightly the process line can only be a little flatter without encroaching on the xe2x80x9cforbidden zonexe2x80x9d and hence the carbon consumption also can only be lowered by a small amount. Of course, this consideration does not exclude the substitution of carbon by other reducing agents.
Thus, in comparison with an autothermic process the higher carbon demand of the blast furnace and the chemical energy present in the furnace gas remain. The diagram shows that the CO/CO2 ratio is 3/2. ⅓of the furnace gas is burnt (afterburning) and after transfer through a regenerator the energy is reintroduced in the form of sensible heat to the combustion air by the tuyeres. ⅔of the furnace gas leave the process with approximately 20% of the energy introduced in the form of coke.
The discussion of the blast furnace method has shown that although precombustion of the carbon is straightforward in technological terms it results in a position of the process line close to the xe2x80x9cforbidden zonexe2x80x9d. Accordingly, the lowest possible slope of the process line and hence the lowest possible carbon consumption cannot be achieved with this method.
The methods can be divided into slow and rapid reduction processes in which the residence time of the charged material amounts to several hours. Processes in a blast furnace and in a rotary-tube furnace in which reduction and combustion to produce heat take place in the same chamber are considered to be slow reduction processes.
The Kinglor-Metor process and the Hoganes process have been proposed as slow autothermic reduction processes. In these processes a precombustion step takes place outside of the reduction chamber. The disadvantage of this method, however, is the low efficiency of the transfer of the heat produced by combustion back into the reduction chamber by thermal conduction. In this case the heat must first of all flow through a ceramic wall and then be conveyed onwards within the reactor by convection in an almost stagnant gas. In another proposal (DE 39 28 415 A1) heat transfer from an external combustion chamber is effected via a heating duct system. On account of this slow type of heat transport the specific efficiencies of these process are low.
Rapid reduction processes are processes in which the residence time of the charged material is less than one hour, operations being conducted at elevated temperatures.
These processes include the Festmet process and the Inmetco process in which combustion takes place in the same chamber as reduction but both reactions proceed far apart from one another. It is intended in this manner that the two gas zones should mix as little as possible. The heat is transferred by radiation from the fire-resistant material in the combustion chamber heated by the combustion reaction into the reduction chamber, there being better heat transfer to the reduction chamber of the coal gasification. On account of the spatial conditions combustion cannot be carried on to complete conversion of CO to CO2. Thus, these processes cannot be autothermic. Considerable excess energy is carried away out of the system.
The aim of the present invention is to provide a method for the extraction of iron by direct reduction of iron ore and a device for this purpose, the intention being that the process should be largely autothermic and accordingly should have a minimum specific coal consumption and energy consumption and that the reduction process should be rapid and capable of being carried out at a high temperature with the result that there is a high level of conversion of ore to metal.
This task is solved according to the invention by the characterising characteristics of the main claim and the coordinated claim in association with their introductory parts.
The use of a separate reduction zone and combustion zone for producing the necessary enthalpy of reduction for the gasification of the coal and the reduction of the iron allows a largely autothermic process in which the lowest possible amount of carbon is employed for reducing the iron. In the combustion chamber the carbon monoxide produced over and above that needed for reduction in the gasification of the coal is burnt and the heat liberated in doing so is transferred via a high-temperature heat exchanger to gas from the reduction chamber. A substantial improvement relative to the state of the art is the forced circulation of gas from the reduction chamber then through the heat exchanger and back into the reduction chamber as a result of which extremely efficient transfer of heat to the coal to be gasified and the iron ore to be reduced is effected. By using a high-temperature heat exchanger, that is a regenerator, in combination with the forced circulation of the reducing gas an autothermic direct reduction process of high specific efficiency is produced with which the throughput time for ores is lowered from hours to minutes.
According to the invention the heat is transferred by rapidly flowing gas to the ore/coal/pellets and, moreover, heat transfer takes place at high temperatures as a result of which the reduction time in the region of the rapid reduction process drops and can be lowered to 10 to 15 minutes. The degree of conversion of the ore to metal rises to over 90%.
By means of the measures specified in the subsidiary claims advantageous refinements and improvements are possible.
For the forced circulation it is suitable to use a fan, for example, which is connected to the heat exchanger and circulates the reducing gases through the heat exchanger. This is particularly advantageous when before they are conveyed through the heat exchanger and the fan the reducing gases are cooled by a cooler to temperatures of less than 300xc2x0 C., for example, so that conventional fans can be used.
It is particularly advantageous to use a plurality of heat exchangers constructed as regenerators through each of which the hot combustion gases flow for a time, for example, and which are heated up after which the reducing gases previously cooled to 300xc2x0 C. are then conveyed through the same regenerator. With the aid of suitable valve connections the heat of the combustion gases can be transferred particularly advantageously to the reducing gases by means of a cyclic sequence of heating and cooling processes of the regenerator or regenerators, only insubstantial quantities of oxygen passing from the waste gases into the reducing gases.
If four regenerators are employed simultaneously the essential heat exchange processes can be carried out synchronously for different spatial regions of the reduction chamber and in this way a particularly effective and rapid reduction of the iron can be achieved. The corresponding four steps for the optimum utilisation of the heat produced in the afterburning of the reducing gases are as follows.
a) Hot gas is drawn in from the combustion chamber by a regenerator and conveyed on to a waste gas unit. In doing so the combustion gas cools and the regenerator absorbs heat.
b) Cooled reducing gas is blown through the heated regenerator and heated to the working temperature of the reduction reactor. This heated reducing gas is blown back into the reducing chamber and in this way carries the required enthalpy of reaction to the coal and the iron ore.
c) This regenerator is cooled further by passing ambient air at room temperature through it and passing it on to the combustion chamber. In doing so the cold ambient air containing oxygen is preheated and this contributes to an improvement in the energy balance.
d) Gas is sucked out of the reduction chamber and through the regenerator which has now cooled almost to ambient temperature as a result of which it is strongly cooled. This cooled gas can then be passed on to a regenerator which carries out the step described under b).
It can immediately be seen that by using four regenerators simultaneously a four-phase process optimised in terms of time and energy can be carried out. Further variations of this method are also conceivable, for example the use of more or fewer regenerators, eg by dispensing with the preheating of the combustion air.
The method according to the invention and the device according to the invention are employed with particular advantage in a rotary-tube furnace.